1. Abstract We propose to create a NIMH P50 ALACRITY Center that develops the collaborations, data systems, and methods necessary to form a state-wide Laboratory for Early Psychosis Research, aka the LEAP Center. While recent trial evidence suggests great promise for coordinated specialty care (CSC) for patients with first episode psychosis (FEP), there remain several gaps in the knowledge base for the care of psychosis in general. Now is an ideal time to address these gaps because of recent policy changes enhancing FEP care financing, practice changes in the field, and methods developed outside of mental health. To create this Center, we will start with interdisciplinary collaborations between CSC clinics in Massachusetts, policy makers/regulators, and stakeholders, and include FEP experts from across the country plus scientific experts from outside of mental health, e.g., in data science, machine learning, epidemiology, and health policy. We also will leverage recent Massachusetts efforts to mandate, standardized, and support data collection on FEP care structures, delivery, and outcomes from all FEP clinics within the state. We also will apply and develop modern methods for clinical prediction and comparative effectiveness research (CER), e.g., machine learning and g-methods, for FEP research. Accordingly, we have three overall Center aims: 1) Collaborations; 2) Data systems; and 3) Prediction and CER methods. The Center will start with three foundational and complementary projects: 1) Using the state All Payer Claims Database (APCD), Project 1 will apply a population-level approach to examine which patients receive care in FEP clinics offering CSC versus elsewhere, as the number of clinics increases, then estimate the treatment effect on unfavorable clinical event rates such as hospitalizations; 2) Project 2 will review the data collected by the state from all of the FEP clinics, assess and improve data collection quality, validate measures, and integrate perspectives; and 3) Project 3 will define clusters of patients using longitudinal outcome data (i.e., begin to unpack the amount of clinical heterogeneity), predict the cluster type for individual patients, and examine the impact of CSC treatment accounting for this clinical heterogeneity. The Administrative Core and Methods Core (Prediction and CER) bind these projects together. The Center will include a group of Scientific Advisors from across North America, Center Faculty from diverse scientific disciplines, including those that historically have had little exposure in mental health research, and Stakeholders, including policy makers, organizational decision makers, and patient and family advocates. All three groups will be deeply involved in the Center from design to dissemination. The overall goal is to create a state-wide learning health system for early psychosis. This effort to create the protocols and apply the methods necessary to convert large amounts of data into useful clinical and policy knowledge will inform other national efforts, e.g., NIMH?s EPINET.